Hum Genet (1984) 68:148-153 © Springer-Verlag 1984 The identification of a DNA polymorphism of the a fibrinogen gene, and the regional assignment of the human fibrinogen genes to 4q26-qter S. E. Humphries 1 , A. M. A. Imam 3 , T. P. Robbins I, M. Cook 1 , B. Carritt 2, C. Ingle 1 , and R. Williamson 1 1 Department of Biochemistry, St. Mary's Hospital Medical School, Paddington, London W2 1PG ~MRC Human Biochemical Genetics Unit, University College, London, Great Britain Summary. We have identified a common restriction fragment length polymorphism of the c~ fibrinogen gene with the enzyme TaqI. This polymorphism is probably due to a single base change that creates or destroys a TaqI recognition site about 1000 basepairs from the 3' end of the c~ fibrinogen g~ne. The frequency of the rare allele in 83 unrelated healthy individuals is 0.33. We have used in situ hybridisation of the a fibrinogen cDNA to localise the gene on chromosome 4@9-31. We have confirmed this regional localisation by restriction fragment detection in a human × Chinese hamster somatic cell hybrid which contains a translocated human chromosome 4 with a breakpoint at 4q26. The a,/3, and 3' fibrinogen genes are all present on human chromosome 4q26-qter. Introduction Fibrinogen (factor I) is an abundant plasma protein which par- ticipates in the final steps of the coagulation pathway in higher vertebrates. A number of genetically determined variants for human fibrinogen have been reported (Bloom 1981) many of which have an increased clotting time, some with pathological consequences. Elevated levels of fibrinogen and other coagu- lation factors have also been correlated with a predisposition to ischaemic heart disease (Mead and North 1977; Mead et al. 1980). Native fibrinogen (tool. wt. 340,000) is a symmetrical tri- nodular molecule composed of dimers of c~,/3, and 3" chains linked by disulphide bridges. The complete amino acid sequences of all three chains have been determined (Blom- back et al. 1976; Lottspeich and Henschen 1977; Doolittle et al. 1979; Watt et al. 1979). The chains show sequence homo- logy suggesting that a common ancesteral sequence under- went duplication and divergence. This relationship has been confirmed recently with the isolation and sequencing of cDNA clones coding for rat and human fibrinogen chains (Crabtree and Kant 1981; Kant et al. 1983a; Rixon et al. 1983; Chung et al. 1983a,b; Imam et al. 1983). The gene for 3"fibrinogen has been shown to be on the long arm of human chromosome 4 by linkage of a 3' chain variant to the MNSs blood group locus (Olaisen et al. 1982). It has recently been demonstrated that in both humans and rats, the fibrinogen genes are physically linked (Kant and Crabtree 3 Present address: Institute for Cancer Research, Fulham Road, London SW3, England Offprint requests to: S. E. Humphries 1983; Kant et al. 1983b) and it would therefore seem likely that in the human genome all three fibrinogen genes are clus- tered on the long arm of chromosome 4 (Kant et al. 1983b). We have confirmed this by in situ hybridisation of the a fibri- nogen cDNA clone and by Southern blot hybridisation to DNA from a hybrid cell line with a translocation involving 4q26-qter. We also report a common restriction fragment length polymorphism (RFLP) of the c~ fibrinogen gene that will be useful for linkage studies and mapping the chromo- some. Materials and methods Gene probes for the fibrinogen chain The c~ and 3' fibrinogen probes (designated pAF1 and pGF1 respectively) have been described previously (Imam et al. 1983). They consist of inserts of 1950bp (pAF1) and 950bp (pGF1) in the PstI site of the vector pKT218, and contain the majority of the coding region and probably all of the 3' untranslated region. The fi fibrinogen probe (pBF5) consists of an insert of 1450bp in pKT218. A partial sequence of this clone is presented in Fig. 1. This confirms that the clone codes for/3 fibrinogen, and from restriction fragment mapping (A. Imam unpublished work) it is likely that the clone contains the majority of the coding region and probably all of the 3' untranslated region. Plasmid DNA was prepared by a modi- fied cleared lysis method (Birnboim and Doly 1979) and purif- ed on CsC1 gradients. 277 Ala Thr Asn Thr Asp Gly Lys CTGCAGGGGG GGGGGGGGGG GGGGGGGGGG T GCA ACC AAC ACA GAT GGG AAG i0 20 30 40 50 PstI 287 297 Asn Tyr Cys Gly Leu Pro Gly Glu Tyr Trp Leu Gly Asn Asp Lys AAT TAC TGT GGC CTA CCA GGT GAA TAT TGG CTT GGAAAT GAT AAA 60 70 80 90 307 Ile Ser Gln Leu Thr Arg Met Gly Pro Thr Glu Leu Leu Ile Glu ATT AGC CAG CTT ACC AGG ATG GGA CCC ACA GAA CTT TTG ATA GAA zoo zlo 12o z3o z4o Fig. 1. Partial nucleotide sequences of the clone pBFI and the corre- sponding amino acid sequences. The numbering of amino acids is as reported in the literature. The recombinant was sequenced by sub- cloning PstI fragments into the PstI site of the M13 rap9 poly linker, as previously described (Imam et al. 1983)